基于EVMD-LNMF的复合故障信号分离方法

PDF(1158 KB)

振动与冲击 ›› 2019, Vol. 38 ›› Issue (16) : 146-152.

论文

王梦阳，王华庆，董方，任帮月，宋浏阳

作者信息 +

A method of compound fault signal separation based on EVMD-LNMF

WANG Mengyang , WANG Huaqing, DONG Fang, REN Bangyue, SONG Liuyang

Author information +

文章历史 +

摘要

非负矩阵分解算法（NMF）和变分模态分解算法（VMD）已用于复合故障信号的分离，但VMD算法过程中模态分量个数难以确定，且NMF算法由于缺少相关约束，对故障源相互耦合，特征信息微弱分解效果并不理想。为此提出了基于能量收敛因子为判据的变分模态分解算法（EVMD）与局部非负矩阵分解算法（LNMF）相结合的复合故障信号分离方法。构造了能量收敛因子，并以此为判断准则，自适应确定VMD算法中模态分量个数；将获得的模态分量重构组成模态矩阵，采用邻近特征值占优法获取LNMF算法中的最优分解维数；对模态分量作LNMF算法处理，突出局部特征信息，从而分离出耦合的多故障信号，提取故障特征信息。仿真及轴承复合故障实验结果表明：提出的基于EVMD-LNMF的信号分离方法，明显优于未改进的VMD-NMF方法，可以有效分离并提取出外圈与滚动体冲击性特征，实现了轴承的复合故障诊断。

Abstract

Non-negative matrix factorization (NMF) and variational mode decomposition (VMD) algorithms had been used to separate compound fault signals; however, the number of decomposition components was difficult to determine in the VMD algorithm.What’s more，due to the lack of related constraints, it was not satisfactory for the NMF algorithm to compound fault and weak feature.A method of compound fault signal separation based on EVMD and LNMF was proposed.Firstly, an energy convergence factor was defined，and used as a criterion to determine the number of decomposition components adaptively.Secondly, the acquired decomposition components were reconstructed as a mode matrix, and the optimal decomposition dimension in the LNMF algorithm was obtained using the method of adjacent eigenvalues domination.Finally, The LNMF algorithm was used to enhance the local characteristic information.The fault feature information was extracted.The simulation and test results show that the proposed method could effectively separate and extract the compound fault features of outer ring and rolling element and realize their fault diagnosis, outperforming the unimproved VMD-NMF approach.

导出引用

王梦阳，王华庆，董方，任帮月，宋浏阳 . 基于EVMD-LNMF的复合故障信号分离方法[J]. 振动与冲击, 2019, 38(16): 146-152

WANG Mengyang,WANG Huaqing, DONG Fang, REN Bangyue, SONG Liuyang. A method of compound fault signal separation based on EVMD-LNMF[J]. Journal of Vibration and Shock, 2019, 38(16): 146-152

参考文献

[1] 刘军伟. 基于盲源分离的轴承复合故障诊断方法研究[D]. 重庆: 重庆大学, 2016.
LIU Junwei. Study on compound fault diagnosis method of rolling bearing based on blind source separation[D]. Chongqi: Chongqing University, 2016.
[2] HAO Yansong, SONG Liuyang, KE Yanliang et al. Diagnosis of compound fault using sparsity promoted-based sparse component analysis[J]. Sensors, 2017, 17, 1307.
[3] 陈晓朦. 单通道盲源分离方法研究及其在抽油机故障诊断中的应用[D]. 西安: 西北工业大学, 2016.
CHEN Xiaomeng. Research on single channel blind source separation and application in fault diagnosis for pumping unit[D]. Xi’an: Northwestern Polytechnical University, 2016.
[4] WANG Jinjiang, Robert Gao, Yan Ruqiang. Integration of EEMD and ICA for wind turbine gearbox diagnosis[J]. Wind Energy, 2014, 17(5): 757-773.
[5] TANG Gang, Luo Ganggang, Zhang Weihua, et al. Underdetermined blind source separation with variational mode decomposition for compound roller bearing fault signals[J]. Sensors, 2016, 16, 897.
[6] Sayeh Mirzaei, Hugo Van Hamme, Yaser Norouzi. Under-determined reverberant audio source separation using Bayesian non-negative matrix factorization[J]. Speech Communication, 2016, 81(16): 129-137.
[7] S.Abdali, B.NaserSharif. Non-negative matrix factorization for speech/music separation using source dependent decomposition rank, temporal continuity term and ﬁltering[J]. Biomedical Signal Processing and Control, 2017, 36(17): 168-175.
[8] F.J.Canadas-Quesada, N.Ruiz-Reyes, J.Carabias, et al. A non-negative matrix factorization approach based on spectro-temporal clustering to extract heart sounds[J]. Applied Acoustics, 2017, 125(10): 7-19.
[9] 唐贵基，王晓龙. 变分模态分解方法及其在滚动轴承早期故障诊断中的应用[J]. 振动工程学报，2016，29(4):638-648.
TANG Guiji，WANG Xiaolong. variational mode decomposition method and its application on incipient fault diagnosis of roller bearing[J]. Journal of Vibration Engineering，2016，29(4):638-648.
[10] 马增强，李亚超，刘政，等. 基于变分模态分解和Teager能量算子的滚动轴承故障特征提取[J]. 振动与冲击，2016，35(13):134-139.
MA Zengqiang，LI Yachao，LIU zheng，et al. Rolling bearings' fault feature extraction based on variational mode decomposition and Teager energy operator[J]. Journal of Vibration And Shock，2016，35(13):134-139.
[11] 刘尚坤，唐贵基，王晓龙. 基于改进变分模态分解的旋转机械故障时频分析方法[J]. 振动工程学报，2016，29(6):1119-1126.
LIU Shangkui，TANG Guiji，WANG Xiaolong. Time frequency analysis method for rotary mechanical fault based on improved variational mode decomposition[J]. Journal of Vibration Engineering，2016，29(6):1119-1126.
[12] Lee D D，Seung H S. Learning the parts of objects by non-negative matrix factorization[J]. Nature，1999，401(6755):788-791.
[13] 牟伟杰，石林锁，蔡艳萍，等. 基于KVMD-PWVD与LNMF的内燃机振动谱图像识别诊断方法[J]. 振动与冲击，2017，36(2):45-51.
MU Weijie，SHI Linsuo，CAI Yanping，et al. IC engine fault diagnosis method based on KVMD-PWVD and LNMF[J]. Journal of Vibration And Shock，2017，36(2):45-51.